The Alkaloids Volume 6

A Review of the Literature Published between July 1974 and June 1975

By M. F. Grundon

The Royal Society of Chemistry

Copyright © 1976 The Chemical Society
All rights reserved.
ISBN: 978-0-85186-307-8

Contents

Chapter 1 Biosynthesis By R. B. Herbert, 1,
Chapter 2 Pyrrolidine, Piperidine, and Pyridine Alkaloids By A. R. Pinder, 54,
Chapter 3 Tropane Alkaloids By G. Fodor, 65,
Chapter 4 The Pyrrolizidine Alkaloids By D. H. G. Crout, 72,
Chapter 5 Indolizidine Alkaloids By J. A. Lamberton, 86,
Chapter 6 The Quinolizidine Alkaloids By M. F. Grundon, 90,
Chapter 7 Quinoline, Quinazoline, and Acridone Alkaloids By M. F. Grundon, 103,
Chapter 8 β-Phenethylamines and the Isoquinoline Alkaloids By N. J. McCorkindale, 110,
Chapter 9 The Aporphinoids By M. Shamma, 170,
Chapter 10 Indole Alkaloids By J. E. Saxton, 189,
Chapter 11 Lycopodium Alkaloids By W. A. Ayer, 252,
Chapter 12 Diterpenoid Alkaloids By S. W. Peletier and S. W. Page, 256,
Chapter 13 Steroidal Alkaloids of the Apocynaceae Buxaceae, Asclepiadaceae, and of the Salamandra–Phyllobates Group By F. Khuong-Huu and R. Goutarel, 272,
Chapter 14 Solanum and Veratrum Steroidal Alkaloids By D. M. Harrison, 285,
Erratum,
Author Index,

CHAPTER 1

Biosynthesis

BY R. B. HERBERT

1 Introduction

The number of papers appearing on alkaloid biosynthesis continues to be large. In reviewing this work the practice of referring to previous Reports is continued and, in order to make this background material more conveniently accessible, these Reports are listed as references 1 — 5. Access to work reported before 1969 is most easily gained through two excellent and comprehensive reviews and reference is made to these where appropriate. The reader is also recommended to consult the alternative annual survey on alkaloid biosynthesis, where the approach is different from the one adopted here.

2 Piperidine, Pyridine, and Pyrrolidine Alkaloids

Piperidine Alkaloids (General). — As part of a study of the role of cadaverine (1) in the biosynthesis of piperidine alkaloids samples of (1), stereospecifically labelled with tritium at C-1, were tested as precursors for N-methylpelletierine (2). These samples were prepared by decarboxylation of L-lysine through the action of L-lysine decarboxylase (isolated from Bacillus cadaveris). Although this reaction was known to proceed stereospecifically it was not known at the time if the result was retention or inversion of configuration. It has recently been shown, however, that the consequence of the enzymic decarboxylation is retention of configuration, and so [1A-3H] cadaverine is (1R)-[1-3H]cadaverine. It follows then, from the earlier conclusion about cadaverine incorporation into N-methylpelletierine (2), that it is the 1-pro-R proton from cadaverine (1) which is retained at C-2 of (2) and the 1-pro-S hydrogen atom which is lost.

Anatabine and Anabasine. — Although the biosynthetic pathway to anabasine (3) has been delineated in detail, little evidence had been obtained until recently on the way in which anatabine (8) is formed: [2-14C]lysine and [2-14C]-4-hydroxylysine were found not to label anatabine and results with 14CO2 indicated that anabasine was not a precursor of anatabine.

In a recent study, [2-14C]lysine was fed to Nicotiana glutinosa, and although it was specifically incorporated into anabasine (3) in the expected manner it did not label the anatabine (8), thus according with the previous result.

Nicotinic acid (6) is well established as a precursor of the pyridine ring of nicotine and is also a precursor of this moiety in anabasine. [6-14C]Nicotinic acid, as might therefore be expected, gave radioactive anatabine, but surprisingly the activity was divided equally between C-6 and C-6′ ([carboxy-14C]nicotinic acid failed to label any of the alkaloids significantly). It follows that nicotinic acid (6) is the source for both rings of anatabine and the equal distribution of label indicates that the two units from which the alkaloid is formed are closely related, if not identical.

The manner in which the two nicotinic acid units may be joined (Scheme 1) is suggested by analogy with the probable intermediacy of a dihydronicotinic acid in nicotine biosynthesis. It is further suggested that anatalline (4) and nicotelline (5) are trimers of the dihydro-derivative (7).

The anabasine obtained after feeding [6-14C]nicotinic acid was found to be labelled almost exclusively in the pyridine ring, so no significant conversion of anatabine (8) into anabasine (3) occurs.

In accord with previous conclusions about the relationship between lysine and anabasine (3), lysine has been found to be a precursor for (3) in Anabasis aphylla along a pathway which does not involve symmetrical intermediates. Aspartic acid was found to serve as a precursor for both rings of anabasine whilst lysine was incorporated into lupinine, again in accord with previous results.

Lobeline. — The results of feeding experiments with DL-[2-14C]lysine and dl-[2-14C]phenylalanine in Lobelia inflata have shown that these amino-acids are both specific precursors for the alkaloid lobeline (13). In further experiments, DL-[3-14C]phenylalanine, [3-14C]cinnamic acid, and [3-14C]-3-hydroxy-3-phenylpropionic acid [as (9)] have been found to be specific precursors for lobeline (13). These results are consistent with the anticipated pathway to lobeline illustrated in Scheme 2, with benzoylacetic acid (10) as the intermediate which couples with Δ-piperideine to give the intermediate (11). The probability of 3-hydroxy-3-phenylpropionic acid (9) being an intermediate in lobeline biosynthesis is increased by the isolation of this acid from L. inflata.

In contrast to the biosynthesis of many other piperidine alkaloids, the incorporation of [2-14C]lysine into lobeline (13) was found to be symmetrical, i.e. C-2 and C-6 were equally labelled. Symmetrization of the label could occur on formation of lobelanine (12), a known late precursor, or at a possible earlier intermediate, cadaverine (1). Cadaverine was found, however, to be a much less efficient precursor for lobeline than lysine or Δ1-piperideine, which argues strongly that it is not an obligatory intermediate in lobeline biosynthesis (cf. the discussion on piperidine alkaloid biosynthesis in ref. 12) and thus symmetrization of lysine label occurs via lobelanine (12).

The biosynthesis of Lobelia alkaloids with C4 units at C-2 and C-6 can be accounted for in terms of the pathway shown in Scheme 3, in which it is envisaged that acetate-derived 3-oxohexanoic acid (14) condenses with Δ1-piperideine to afford, after methylation, the keto-amine (15) which is oxidized and condensed with a further molecule of 3-oxohexanoic acid (14). Truncation of the (16) formed affords the alkaloid (17). Evidence for the validity of this pathway has been adduced from the results of feeding [214-C]lysine and [1-14C]acetate. The former precursor provides the piperidine ring of (17) as expected and acetate the side chains, with one half of the activity on the carbonyl groups as required by the hypothesis.

Santiaguine. — The origin of the curious a-truxillic acid (18) moiety of santiaguine (21), an alkaloid of Adenocarpus species, has been shown to be cinnamic acid. This moiety arises, the available evidence has suggested, during the biosynthesis of santiaguine as a result of the dimerization of adenocarpine (20). An alternative route to santiaguine, which involves the dimerization of cinnamic acid, followed by condensation of the α-truxillic acid formed with tetrahydroanabasine (19), had not been explored, however. Recent results demonstrate that α-truxillic acid (18) exists free in Adenocarpus foliosus, is specifically labelled by cinnamic acid, and is itself specifically incorporated into santiaguine (21). Thus two distinct pathways to santiaguine may operate (Scheme 4) but it appears, from the relative incorporation efficiencies, that the route via adenocarpine (20) is, at least, the major one.

α-Truxillic acid (18) can be formed photochemically in vitro, from cinnamic acid, but it seems that such a route is not operative in vivo, for no significant difference in the incorporation of labelled adenocarpine (20) into santiaguine (21) could be observed with plants grown in light or darkness.

The origins of the tetrahydroanabasine (19) moieties of santiaguine have been examined in feeding experiments with DL-[2-14C]- and DL-[6-C]-lysine. Tracer was incorporated specifically and unsymmetrically into each of the heterocyclic rings (Scheme 5). Labelled Δ1-piperideine (22) was incorporated in like manner and is thus a likely intermediate in the formation of each piperidine ring (Scheme 5). These results are in accord with the carefully delineated pattern of piperidine alkaloid biosynthesis. Application of the model developed to explain the incorporation of lysine and cadaverine into these alkaloids allows one to include cadaverine as a precursor for santiaguine, as the tracer results suggest. [The generation of the tetrahydroanabasine skeleton of (21) from lysine makes a notable contrast with the formation of anatabine (8) from nicotinic acid; see above].

The above experiments have thrown up a point of practical interest and possible application elsewhere. When radioactive Δ1-piperideine was fed to A. foliosus followed by labelled cinnamic acid, the incorporation of the latter was 10 times as high as previously found. This was tentatively attributed to stimulation of alkaloid biosynthesis by the presence of both precursors necessary for alkaloid synthesis.

Quinolizidine Alkaloids. — Previous results demonstrate that the quinolizidine skeleton in its entirety derives from lysine. Further research has indicated that lysine is a precursor of all the alkaloids of this type in five species of Leguminosae. From the levels of activity observed in the individual alkaloids it was concluded that saturated alkaloids are precursors for those with a pyridone ring. This was supported by the observation that label from radioactive sparteine (24) and lupanine (25) appeared in more highly oxidized alkaloids. (This compares with a similar situation in the biosynthesis of matrine-type alkaloids.) A metabolic grid for the biosynthesis of quinolizidine alkaloids from lysine was proposed, based on these results, other more convincing evidence, and the taxonomical distribution of quinolizidine alkaloids.

In young Lupinus nanus plants added cadaverine but not putrescine or hex-amethylenediamine lowered the incorporation of lysine into thermopsine (26). (In flowering plants the difference between cadaverine and the other two diamines was less marked.) These results provide further evidence for the intermediacy of cadaverine in quinolizidine biosynthesis.

N-Methylconiine. — Feeding experiments with L-[Me-14C]methionine in Conium maculatum varieties have established that the methyl group of methionine serves as the source of the N-methyl group of methylconiine (27). This adds further to the picture of coniine biosynthesis and provides an additional example of transmethylation in plants. In further experiments, an enzyme was isolated from C. maculatum that was capable of catalysing the transfer of a methyl group from S-adenosylmethionine to coniine (28), with the formation of N-methylconiine (27). The enzyme was partially purified and characterized in part.

Ricinine. — Quinolinic acid (29) is known to be a precursor of ricinine (32). The results of other studies have indicated that the pyridine nucleotide cycle (Scheme 6) may be a necessary part of ricinine biosynthesis, although the observed increase in incorporation of label into ricinine from radioactive quinolinic acid in the presence of excess exogenous NAD+ has been interpreted as indicating that conversion of quinolinic acid (29) into ricinine occurs by a separate pathway. In order to clarify the relationship of the pyridine nucleotide cycle (Scheme 6) to ricinine biosynthesis the effect of various inhibitors on quinolinic acid incorporation into ricinine and pyridine nucleotide cycle intermediates has been examined.

Azaserine, a glutamine antagonist, is known to inhibit the NAD+ synthetase reaction in which nicotinic acid adenine dinucleotide is converted into NAD+ with glutamine or ammonia as the nitrogen donor. When azaserine or azaleucine was fed to Ricinus communis plants followed by [6-14C]quinolinic acid, a marked decrease in incorporation of radioactivity into ricinine was observed with azaleucine, less so with azaserine. Both azaserine and azaleucine were found also to inhibit the incorporation of [6-14C]quinolinic acid into pyridine nucleotide cycle intermediates [in the case of azaserine the conversion of nicotinic acid dinucleotide into nicotinamide adenine dinucleotide (NAD+) was apparently inhibited].

Ricinine and ethionine were found to inhibit ricinine biosynthesis with shunting of radioactivity from [6-14C]quinolinic acid through the pyridine nucleotide cycle into N-methylnicotinic acid (30) and N-methylnicotinamide (31) (cf. Scheme 6), and quinolinic acid consumption was reduced.

These results strongly indicate a dependency between ricinine biosynthesis and the pyridine nucleotide cycle. The similarity in incorporation of members of this cycle into ricinine is explained by allowing each member to be diverted directly into a pathway leading to ricinine; this explanation covers the observation that excess exogenous NAD+ increases the incorporation of quinolinic acid into ricinine since this precursor can simply be shunted along one of these diversions if the formation of NAD+ is blocked by its presence in large excess.

Tropane Alkaloids. — An abundance of evidence points to phenylalanine as an intermediate in the biosynthesis of the tropic acid moiety [as (34)] of tropane alkaloids like atropine (33), but support has also been obtained for alternative precursors for this moiety. All nine of the carbon atoms of phenylalanine are incorporated into tropic acid (Scheme 7) in a rearrangement reaction which manifestly involves the migration of the carboxy-group rather than the phenyl group. The mechanism of the rearrangement is unknown, but it is circumscribed to some extent by the results of other experiments.

Phcnylpyruvic acid (36) has been found to be a similarly efficient precursor for tropic acid as phenylalanine, but the ready interconversion of phenylalanine and (36) in vivo makes it difficult to say which is closer biosynthetically to tropic acid. Phenyl-lactic acid (37), as it appears in the alkaloid littorine (35), is known to be derived from phenylalanine, and it was found to be a better precursor than phenylalanine for the tropic acid moiety of tropane alkaloids.

Cinnamic acid, a metabolite of phenylalanine, has been linked with tropic acid biosynthesis via its epoxide (38) on chemical grounds, but neither of these compounds was found to act as a tropic acid precursor. That rearrangement was not taking place after linking cinnamic acid to a tropine residue was demonstrated when it was found that cinnamoyltropine (40; 14C labels as shown) was incorporated into (33) with retention only of the tropine label, which is consistent with hydrolysis of (40) to tropine (39) and cinnamic acid in the plant, the latter not being incorporated. Moreover, whilst littorine (35; 14C and 3H labels as shown) was incorporated specifically into atropine (33), the change in isotope ratio demonstrated that incorporation involved prior hydrolysis of littorine to tropine (39) and phenyl-lactic acid; no facile interconversion between tropine, tropic acid, and littorine was apparent because the littorine isolated at the end of the experiment showed the same isotope ratio as the material fed. Thus rearrangement of a tropine ester of lactic acid is not involved in the rearrangement to give tropic acid either.

In Datura species the tigloyl moiety [as (41)] of e.g. meteloidine (42) is known to arise from L-isoleucine via 2-methylbutanoic acid, and although angelic acid (44) is also derived from L-isoleucine it is not a precursor for tropane alkaloids in Datura innoxia, and so no cis-trans isomerization of (44) to (41) occurs in this plant. In further experiments L-[U-14C] isoleucine has been found to give radioactive tigloidine (45) and 3α-tigloyloxytropane (43) in Physalisperuviana and radioactive (46) in D. ceratocaula; in the case of (46) the radioactivity was shown to be confined to the 2-methylbutanoyl moiety, which notably is of the same configuration as L-isoleucine.

Pyrrolizidine Alkaloids. — The C10 necic acids of the senecic acid type, which appear as diesters in alkaloids like senecionine (47), have been shown to be formed from two units of L-isoleucine as illustrated in Scheme 8. Further experiments have shown that, of the four stereoisomers of isoleucine, only L-isoleucine is used efficiently for senecionine biosynthesis in Senecio magnificus plants (of the other isomers only L-alloisoleucine is significantly incorporated, but it has still to be established whether this incorporation is specific or not). This selection of one particular isomer of isoleucine for biosynthesis contrasts with the finding that certain micro-organisms can use the different stereoisomers of this amino-acid for growth and biosynthesis.

Acetate has been reported to be a precursor for the necic acid component, monocrotalic acid [as (48)], of monocrotaline (49). Rigorous attempts to confirm this rinding have met only with failure. Although acetate was rapidly taken up into the plant it was not incorporated into (49). The earlier conclusion, it was suggested, may have been occasioned by working with impure alkaloid.

(Continues…)Excerpted from The Alkaloids Volume 6 by M. F. Grundon. Copyright © 1976 The Chemical Society. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.